Audio & video
Bring media into one transcript review flow.
THE LOCAL AUDIO WORKSPACE
Give each character a voice. Refine a single line. Finish the whole story. Speech, transcription and voice assets come together on your Mac.
Script / 03 passages
The station clock had stopped at eleven.
We still have time.
Then let’s take the long way home.
Character → Voice
Passage order illustration
Start with structure. Narration and dialogue each get their own passage.
Interactive workflow illustration · No audio or generation
01 / CREATE
A short narration, a scene between characters, or an entire book. Choose a structure, then make it sound like yours.
Work with readable passages. Refine a section and check its place in the whole, without starting the entire piece again.
The last train home
The station clock had stopped at eleven. Across the tracks, a single café was still open, its windows catching the last of the rain.
Mara folded the ticket into her coat pocket. There was no hurry now. For once, the evening belonged to them.
At the next crossing, they turned away from the station. The city was quieter here, and the long way home felt like the right one.
Keep character roles separate from voice choices. Change the voice of one character while keeping the script in place.
Mara Warm storyteller
MaraWarm storyteller
“We still have time,” she said.
Narrator
Beyond the platform, the city kept moving.
Eli
“Then let’s take the long way home.”
Import EPUB, TXT or Markdown. Organize chapters, generate in segments, review the results, and bring the finished book together for export.
Choose a chapter to explore its structure. This example does not run generation jobs.
The station clock had stopped at eleven.
Interactive example · Fixed content, no live generation
02 / TRANSCRIBE & REVIEW
Start with audio or video. Find a line in time, correct the words, then carry your current draft into subtitles, a document or the next production step.
Edits update only this subtitle line. Example timings stay fixed.
1 00:12,400 --> 00:16,800 We’ll meet at the peer before sunset. 2 00:17,200 --> 00:20,400 Bring the old map with you. 3 00:20,800 --> 00:24,000 We can take the long way home.Download example SRT ↓
Interactive example · Edit preset text and download a text file. No transcription runs and no media audio changes.
Bring media into one transcript review flow.
Locate, edit and check in one place.
TXT / SRT / VTT / JSON for the next step.
03 / VOICE ASSETS
Describe a voice, or bring an authorized recording. Compare candidates, save a voice asset, and use it across narration, dialogue and Agent workflows.
Three starting points. One voice library.
For example: a bright, natural young voice, unhurried and clear, telling a story on an early morning walk.
Compare and save generated candidates in the App. The cards below are examples.Prepare a clear recording you have permission to use. Check the reference text, then save the voice to your library.
Reference cloning is a separate route from voice design. Confirm your rights to the recording.Browse compatible factory and saved voices. Choose the model and language, then cast your narration or dialogue.
Voices must match the model and language. Every voice is not compatible with every model.Selected voice · Example
Light and clear, with the ease of a story just beginning.
Single voice · Character dialogue · Agent calls
Bring your selection into the character script above. This example has no audio preview.
Selecting a candidate updates this fixed example. Voice design and cloning happen in the App; use only recordings you are authorized to use. Interactive example · Fixed content, no live generation
04 / KEEP GOING
Keep the surrounding work in reach. Models, jobs, history and exports each have a place, so your next step stays connected to the project.
Choose by task. Check download and readiness states before starting your next generation.
Explore modelsFollow generation, transcription and downloads in one place, even after leaving a workspace.
About jobsReturn to finished results, find their voices and source material, and keep refining.
History & filesImport EPUB, TXT or Markdown. Review chapters, generate and check them, then export the book.
Audiobook workflowReview recordings or video with transcript and silence candidates. Confirm edits before exporting media.
Media editingPrepared speech models infer on your Mac. External text services are yours to configure and choose.
Privacy & connections05 / AGENTS
MCP, Skills and the local API share the same voices, jobs and artifacts as the desktop App. Let an Agent handle the steps; keep the creative decisions yours.
Explore Agent accessYOUR WORKSPACE
Light, dark and six accent colors. Explore a small web preview below; choose your actual workspace appearance in the App.
The station clock had stopped at eleven.
SONANO
Find your workflow. Then give it a voice.